Human Health Data Visualization Based On Sensors

In recent years, concept of health medicine has promoted the development of visualization technology, which is widely used in the fields of medical and human health information, user-oriented health data visualization service attracts more and more attention from researchers. Visualization is a technique to convert abstract data into graphics or images displayed on the screen and interactive processing. With the development of virtual reality technology, abstract data can be displayed as a three-dimensional model on the computer screen, user-friendly interactive process has enriched the visual expression which is easy for users to handle interaction and expend visual expression. Human health data visualization consists of modeling and graphing. This study conducted a detailed implementation based on joint angle data from human daily activities. The contents of this paper is organized as follows.Firstly, according to the joint angle data, a set of human motion model and coordinate transformation were established with respect to the measurement requirement. Assisted by the sensors to get the body motion angle, virtual model of reproduce three-dimensional body posture was represented on PC interface through visualization techniques to facilitate the monitoring on user’s daily activities. In addition, a brief overview of data processing and analysis based on the angle data were given out, and characteristics of daily activities were extracted for classification, the movement conditions were evaluated by analyzing the correlation parameters calculated from activities.Secondly, ZigBee sensor network technology was selected in this study, MPU6050 sensors and CC2530 BSN system were designed and implemented to obtain the multi-joint angle data of the human body movement, including hardware design and system software design, and dressing method was implemented to acquire multiple human joints angle data.Finally, based on BSN data acquisition system, visualization software platform was carried out, including visual interface based on MFC and data analysis and processing based on Matlab. Visual interface contained data reception module, writing and query the database module, chart display module and OpenGL display module in the virtual environment which is designed to reproduce body posture in real time by a virtual human skeleton model. While in data processing, BP neural network was used to classify human daily exercise, using features of angle data from daily activity monitoring, simultaneously the walking parameters were calculated to identify human movement situation.The synchronous test and experiments were carried out, including performance test on both BSN data acquisition system and PC visualization software platform. The experimental results indicated that sensor nodes can collect angle data accurately in real-time and network communications was of good performance, they can establish a stable and accurate communication with host computer. In data processing, classification and walking distance experiments were carried out. In visual interface, the interactive features, three-dimensional model transformation and each module display features were tested and results showed that the overall system can achieve dynamic interactive visualization of angle data.